---
id: 20260731-T0-15
title: "EvoPINN：AI自动发现物理信息神经网络执行算法"
title_en: "EvoPINN: AI Discovers Executable Algorithms for Physics-Informed Neural Networks"
url: https://ai.daily.yangsir.net/daily/20260731-T0-15
issue_date: 2026-07-31
publish_date: 2026-07-30T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2607.26490
---

# EvoPINN：AI自动发现物理信息神经网络执行算法

arXiv新论文提出EvoPINN方法，利用AI Agent自动搜索可执行的算法来优化物理信息神经网络（PINN）的表示、损失函数和架构，避免人工试错。该方法有望加速偏微分方程求解。

## English Version

**EvoPINN: AI Discovers Executable Algorithms for Physics-Informed Neural Networks**

EvoPINN uses an AI agent to automatically discover executable algorithms for PINNs, optimizing neural representations, loss formulations, and architecture without manual trial-and-error. It speeds up PDE solving tasks.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2607.26490)

**详情页**：https://ai.daily.yangsir.net/daily/20260731-T0-15

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*